© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 44185
    Prev
    Next
    INDEX
    Explanations

    .

    np_max-act · gemini-2.0-flash

    This neuron flags the running‐text fact-and-argument passages of legal opinions (the substantive narrative) rather than the document headers or metadata.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Lazy
    -0.07
     Township
    -0.07
    ='_
    -0.06
    цями
    -0.06
     xf
    -0.06
     apa
    -0.06
    _Page
    -0.06
     Arabia
    -0.06
     navy
    -0.06
     HI
    -0.06
    POSITIVE LOGITS
     최고
    0.07
    	cerr
    0.07
     Meteor
    0.07
    .dynamic
    0.07
    sciously
    0.06
    ValidateAntiForgeryToken
    0.06
    .RowStyles
    0.06
    .normalized
    0.06
    	IN
    0.06
    stringstream
    0.06
    Activations Density 0.028%

    No Known Activations